Computational intelligence for decision-making systems

✍ Scribed by Nikhil R. Pal; Rajani K. Mudi

• Publisher: John Wiley and Sons
• Year: 2003
• Tongue: English
• Weight: 46 KB
• Volume: 18
• Category: Article
• ISSN: 0884-8173
• DOI: 10.1002/int.10098

✦ Synopsis

There have been several attempts to define computational intelligence (CI). Bezdek, 1 in 1994, was the first to introduce this term: " . . . A system is computationally intelligent when it: deals with only numerical (low-level) data, has a pattern recognition component, does not use knowledge in the AI sense; and additionally when it (begins to) exhibit (i) computational adaptivity; (ii) computational fault tolerance; (iii) speed approaching human-like turnaround, and (iv) error rates that approximate human performance."

In 1995, Fogel 2 summarized CI as " . . . These technologies of neural, fuzzy, and evolutionary systems were brought together under the rubric of Computational Intelligence, a relatively new term offered to generally describe methods of computation that can be used to adapt solutions to new problems and do not rely on explicit human knowledge."

Irrespective of the way CI is defined, CI tools should have sufficient ability to solve practical problems. It should be able to learn from experience and be capable of self-organizing. Consequently, it is reasonable to assume that the major constituents of a CI system are artificial neural networks, fuzzy sets, rough sets, evolutionary computation, and immunocomputing.

The major tools of CI also are encompassed by another widely used term, soft computing, which refers to a collection containing neural networks, fuzzy logic, evolutionary computation, etc. Soft computing is defined as a consortium of different computing tools that can exploit our tolerance for imprecision and uncertainty to achieve tractability, robustness, and low cost. 3 We prefer the term "computational intelligence."

An "intelligent decision-making system" should have the capability of dealing with imprecise and uncertain information, learning from examples, deducing from